More information on Spitzer operations

It might seem, if you're just using Spitzer data (as opposed to planning observations), that you don't need to know much about Spitzer operations. But, it turns out that you do need to know a little bit about how Spitzer works in order to understand why the data look the way they do.

Spitzer is a robot

Spitzer is a robot, designed to operate autonomously for 12-24 hours at a time. We at the SSC prepare schedules for the telescope, and this is done for a whole week at once. We then upload the schedule (in whole or in part) up to the spacecraft. The telescope then goes about taking data all by itself. Every 12 or 24 hours (whatever we've told it to do), it "calls home", downloads the data and gets any new instructions. We use the Deep Space Network (DSN) to communicate with Spitzer; pretty much every other spacecraft not in Earth orbit is also competing for this limited resource. We have to plan out well in advance when we will be able to use certain radio dishes that are part of the DSN. For example, when there is a special event, such as an "orbital insertion" (when a spacecraft goes into orbit around another planet), then that event takes priority over all others, that spacecraft gets whatever antennas it needs, and we have to rearrange our uplink/downlink times accordingly.

This operational constraint is why we can't drop everything and make an observation of whatever you want right this second. This also means that there is NOT someone sitting with a joystick (or joystick-like-thing) at the SSC or at JPL, commanding the spacecraft in real time. All of the observations have to be planned in great detail at least 2 months in advance, usually more in advance than that. This is one way that observing from space is very different than observing from the ground.

Spitzer sees (in) donuts

Spitzer is more or less cylindrical in shape. (For pictures of the spacecraft and its components, visit this online gallery.) Its solar panels cover essentially one whole side of the cylinder, and that side must always face the Sun, not only because it is good to keep the spacecraft powered, but because the solar panels help keep the telescope shaded from the Sun, and we need a cold telescope to be able to observe. So the only way in which Spitzer can move is to rotate in a circle with that side of the cylinder always facing the Sun. In this fashion, Spitzer can see a circle, or torus, or donut, of the sky at any one time. ( This cartoon is meant to show this torus.) As Spitzer continues in its orbit around the Sun, it can see the sky directly above and below the orbital plane (the north and south ecliptic poles) all the time, but it can only see objects in the ecliptic plane (the plane of the Solar System - not necessarily just things in the Solar System, but things behind that plane) for a little while - two periods of about 40 days in duration each year. So, over 6 months, Spitzer can see any part of the entire sky. But it can't just point anywhere at any time.

Spitzer has (had) many instruments

Spitzer has three different science instruments on board: IRAC, the InfraRed Array Camera, IRS, the InfraRed Spectrograph, and MIPS, the Multiband Imaging Photometer for Spitzer. (Here is the scientist-focused SSC website which includes detailed information on each of the instruments.)

IRAC has four cameras inside that image using 4 broad-band filters, centered at 3.6, 4.5, 5.8, and 8 microns.

MIPS has three cameras inside that image using 3 broad-band filters, centered at 24, 70, and 160 microns.

IRS is designed to take spectra, and it has two modules inside that take low-resolution (lambda/delta lambda ~64-128) spectra over two wavelength ranges (5.2-14.7 microns and 14.3-35.1 microns), and two more modules that take high-resolution (lambda/delta lambda~600) spectra over two other wavelength ranges (9.9-19.5 microns and 18.9-37.0 microns). In order to help it center on the source of interest, it also has two small imaging cameras, centered roughly on 15 and 24 microns.

All three instruments operated when we had cryogen, from 2003-2009. Starting in Spring 2009, we ran out of cryogen. The telescope passively cools to about 30 K, which allows us to continue to use the shortest two channels of IRAC, but nothing else. NASA funding has permitted us to go 2 years past end of cryogen, and as of Spring 2010, has recently permitted us to continue for at least 1 more year.

Spitzer (sort of) does only one thing at a time

When we had cryogen ...

Only one of the instruments is on at a time, so that means that you can't do simultaneous observations with, say, MIPS and IRAC. However, when one of them is on, for the most part, all pieces of it are on at the same time, and you can't turn a piece off. This still doesn't mean that you can do simultaneous observations using different pieces of a given instrument, for the following reason.

There is only one moving part on the whole spacecraft (and that is the MIPS scan mirror). We do not have, as might be found on ground-based telescopes, a single field of view with, say, a pick-off mirror that rotates to feed light to different instruments. Instead, for a given pointing of the spacecraft, each camera within each instrument sees a slightly different piece of the sky. This figure has a picture of Spitzer's focal plane, e.g., all those fields of view projected onto the sky.

So, what does this mean in practice? Let's walk through an IRAC observation. IRAC has four cameras, and two fields of view (the two large blue squares in the figure linked above). Each one of IRAC's fields of view has a special kind of optics component called a beamsplitter, which reflects some wavelengths of light and allows other wavelengths to pass through. One IRAC field of view is shared by the 3.6 micron and 5.8 micron cameras (IRAC channels 1 and 3) -- the beamsplitter sends the 3.6 micron photons one way and the 5.8 micron photons another way. The other IRAC field of view is shared by the 4.5 and 8 micron cameras (IRAC channels 2 and 4). If you construct an observation that asks the spacecraft to observe a certain target in just 3.6 microns, then the telescope will slew to center that target in the 3.6 micron field of view. Because you can't turn off the other three IRAC cameras, you will also get data in those other three channels "for free." The 5.8 image will also be centered on the target you asked for, because it sees the same target as the 3.6 micron camera. The 4.5 and 8 micron cameras both see a different field of view, offset from the target you asked for. The edges of the two IRAC fields of view are separated by about 1.5 arcminutes. If you asked for Spitzer to observe a single target in 3.6 and 4.5 microns, then the spacecraft would first slew to position your target in the 3.6 micron field of view, take that data, and then slew to put your target in the 4.5 micron field of view and take data there. You would get data in the other two IRAC channels (5.8 and 8 microns) as well, "for free."

For more advanced readers: The direction of the offset of the second IRAC field of view is set by the time and location (in the sky) of the observation -- for the reasons given above ("Spitzer sees (in) donuts"), the focal plane rotates on the sky. This rotation is about a degree a day on the ecliptic poles, and there is no (or very little) rotation on the ecliptic plane. So you can't obtain just any angle at just any time. If you want to be sure a certain region is covered in all four IRAC bands, you need to make a small map that ensures that the region you want to be covered is covered by both IRAC fields of view, regardless of when the data are taken (regardless of rotation angle). Spot helps you do this.

Now that we no longer have cryogen, it's just IRAC that is on, all the time.

Why are data not available immediately after they have been obtained?

Well, the first part of the answer is, as it says above, Spitzer operates autonomously for 12-24 hours at a time, so your data may not even be on the ground for a day after they have been taken.

Another part of the answer comes from the fact that Spitzer operates in "campaigns." For example, IRAC is on, taking data, for days to weeks at a time. Calibration activities are carried out at least at the start and end of the campaign. For the best calibration of your data, we wait until the end of the campaign to reprocess everything. It takes time to do this and to move data around between JPL and the SSC, and through our pipelines here at the SSC. The last step is to make the data available through Leopard. (More information for astronomers on this process.) Spitzer makes a lot of data very quickly -- MIPS and IRAC generate about 20 GB per day (!) -- and just moving these data around takes time!

The last part of the answer may come from something called proprietary periods. Some data have no proprietary period, e.g., they are available to everyone as soon as they make it into the archive. Certain other professional astronomer observations have similar rules. But it is more common for the astronomer who planned the observation to get a little time, usually a year (sometimes less) to be the only one allowed to see the data. After that, the data are public, and anyone can see them.

So, data obtained today will not be available to anyone for up to a few weeks as they make it through our pipelines. After that, the data may be available just to the astronomer (or the team of astronomers) who planned the observation for up to a year. And then, after all of that, you can get to the data. The date that the observation was obtained and the date at which it becomes public are both listed in the archive.